Remote Sensing of Ecosystem Condition and Resilience

Science Center Objects

Ecosystem condition tends to be highly dynamic in response to natural variability in climate, extreme climate events, disturbance events, and human land use activities. Satellite imagery provides a powerful tool to enhance our understanding of ecosystem change at a landscape scale. This research integrates diverse sources of satellite imagery with ancillary datasets to explore how ecosystems respond to and recover from mechanisms of change. Current research is focused on (1) change to wetland and riparian corridor condition and extent, and (2) characterizing patterns in post-fire condition and recovery.  

Why this Research is Important:

Extreme events like droughts, floods, and fires have large impacts on the health and safety of nearby communities. We use data from satellites to help us understand how these extreme events produce changes in water quantity, quality and distribution, as well as the extent of fire damage and post-fire revegetation, information that can help us manage natural systems and protect communities. The analysis of satellite imagery enables us to provide cutting edge science in a cost-efficient manner to both support the needs of local communities and scale easily to support national or global priorities. 

Specific Projects:

  1. The role of land use changes on riparian condition – Agricultural irrigation is a primary use of water consumption nationally. Changes to irrigation can be expected to impact the distribution of water both locally and at a watershed-scale. We are using multi-decadal Landsat imagery with climate datasets and stream gage data to explore how riparian condition responds to patterns in climate as well as changes in irrigation technique. We are using the Missouri River Headwaters Basin in Montana as our study area.
  2. Characterizing anthropogenic wetland loss – To date, most efforts to track wetland loss rely on changes in surface water extent. The goal of this project is to separate temporary wetland loss, attributable to a drought event, from permanent wetland loss, attributable to changes in land use. Our approach is to separately track bare soil disturbance and surface water extent using multi-decadal Landsat imagery across the Mid-Atlantic States. 
  3. Patterns and rates of post-fire conifer regeneration – As fires have become larger and more severe in recent decades, interest has risen in understanding and predicting potential shifts in the successional trajectory of forests following a fire. Changes in species composition or a state transition to a non-forest ecosystem post-fire can have long-term impacts on nutrient, energy and water cycling. We are tracking the post-fire condition of conifer forests under both snow-cover and growing season conditions across the western United States. By pairing the rate of regeneration with climate, topography and burn characteristics we can improve our predictions of future forest condition.    
  4. Enhancing efforts to track wetland fires – detecting fire events across the southeastern United States with multispectral data is particularly challenging due to (1) frequent cloud cover, (2) rapid rates of vegetation growth, and (3) high frequency of lower severity prescribed fires. This project is exploring approaches to improve our remote detection of wetland fires by exploring new sensors as well as by considering non-traditional post-fire spectral trajectories. 

Collaboration:

Collaborations with research partners are a key component of this project. Currently we are actively collaborating with partners from U.S. Environmental Protection Agency (EPA) Region 3, U.S. EPA Region 8, U.S. EPA Office of Research and Development, U.S. Fish & Wildlife Service, non-profit organizations, and Tall Timbers Research Station.